Hacker Newsnew | past | comments | ask | show | jobs | submit | Joel_Mckay's commentslogin

Design for Manufacturability (DFM) will determine if they generate a viable product. Hardware is difficult, with a success rate around 1:66 for a startup.

It is a good idea to consult with a contract manufacturer engineering team early, and stop wasting peoples time including their own.

I have also seen people re-brand China pad-printed generic products, and claim they needed investor help buying a shipping container full of cheap stuff or outright e-waste. Careful, especially if you see multiple versions or the same product pop up at the same time. Best of luck =3


LM555 or 300MB nodejs App on a $300 SoC... the choice is clear... lol =3

NP-hard problems are still NP-hard even with magical LLM pixie sprinkles. =3

If the magic LLM pixie sprinkles can brute force their way to a solution using viable heuristics and their training data, I will use them.

NP-Hard is NP-hard... "heuristics" in this case is just a euphemism for a fantasy cost-function to scam shareholders.

It is a named problem, and can't be misappropriated like the $9Tn worth of FOSS code "AI" scraped and compacted for isomorphic plagiarism tokens. =3


The best solution might be NP-hard. However we don't need the best, we need one that is good enough, and that is more doable.

So, then how are today's pcb layout techs solving these Np-hard problems in your mind, armed only with Altium?

Simple: we aren't.

We can find local optima, but there's no real way to prove a given design is the best of all possible designs. We can only find designs which work and meet the given criteria.

One can prove that a design is correct, but it requires the same class of compute as an LLM to properly simulate. Circuit boards are hellishly complicated once you start simulating EM and RF responses.

As for how a good engineer does this, it's the same way pilots can fly helicopters: intuition. It's a skill built up from lots of reading the rules and theory, and lots of practice. One gets a feel for how the various fields and energies are moving.

Which, when you think about it, is just about how neural nets learn, isn't that neat?

I think the real answer here is that machine learning is absolutely capable of producing satisfactorily correct circuit boards. In principle. But I think the problem space is far too large for today's ML techniques and verification/iteration is just way too expensive. Maybe in another few years, I just don't think we have enough compute for it yet.


Only decades of standards compliance, workmanship training, and experience. Human beings are very good at spotting patterns in noisy jungles, and making rational design tradeoffs.

Auto-routers have been around far longer than even smartphones. Somewhat functional in the trivial problem domains, and always useless where design choices mattered.

Best of luck solving a named problem. =3


Computationally what’s the difference between a human and an LLM besides scale and speed? Trained humans are still using heuristics and shortcuts. Just because these are subconscious and only possessed by talented, trained professionals doesn’t make the skill computationally special.

LLM do not think because they are not real "AI", but it does copy the linear patterns people exhibit if statistically salient within the granularity of the higher dimensional vector search space proximity. Watermarking does skew the compaction slightly, but not far from resulting output patterns. =3

https://en.wikipedia.org/wiki/The_Subservient_Chicken


I do think that slightly smarter LLMs and a good (although not perfect) autorouting algorithm could solve 99% of hobbyist's projects and simple industry boards, due to the repetition of those patterns in real life.

The LLM would learn the heiristics (example: data lines first, power lines later, etc.) and would request the autorouter to do that routing, then take the image output and request a different part (depth-first). If later no routing is posible with that configuration, after some retries it could try another way.

It wouldn't solve complex boards, and engineers are always needed for short comings (and even if not, for research), but I wouldn't say this is something different than poetry, music or pixel art, LLMs can imitate although most of the times in a uncanny way.


What were the chicken's thoughts on the Jacobian conjecture?

> Somewhat functional in the trivial problem domains, and always useless where design choices mattered.

This is laughable when you consider that semiconductor design is the least "trivial" problem domain, and it happens entirely in RTL. There's basically zero human intervention when it comes to layout: even floorplanning is being automated nowadays.


> Only decades of standards compliance, workmanship training, and experience

If that could solve NP-hard problems, computer science would be a very different field.


Magical LLM pixie sprinkles ... is that where magic smoke comes from?

Magic smoke is important, as it indicates a design parameter was exceeded.

I often recommend folks get a mini thermal camera for their phone... as inspecting for issues in electrical, hydraulic, or mechanical systems gets a fair bit easier/safer. =3


They're also NP-hard for the humans doing the majority of the work currently, meaning that good enough heuristics dominate as it is.

Every trace on a PCB is a traveling salesman problem with ballooning complexity.

https://en.wikipedia.org/wiki/Travelling_salesman_problem

While physics informed models do exist, they are still going to burn a lot of compute to generate failure modes people didn't know were possible. =3

https://www.youtube.com/watch?v=T4Upf_B9RLQ


Well, no, it's not.

If anything, it's closer to the bin packing problem. The cost function isn't expressed in terms of lowest cost between nodes, but instead fitting all of the traces with the lowest board layer count while still meeting stuff like EMI crosstalk guidelines. Then there'll be some traces that are very constrained (think DRAM or SERDES links) that have to be length matched, but those are pretty much invariably point to point links that already are heavily automated.

Traces being, say 20% longer than they need to be for an optimal solution doesn't really change the performance of the board in the vast majority of cases since they'll be some of the lowest resistance components of the netlist anyway, as well as the fact that the PCB is normally oversized for the number of traces needed as it fulfills structural/mechanical needs as well as the netlist needs.

And on top of that, because the optimal solution probably is NP-hard, once again, the industry doesn't look for optimal, only good enough. If layout techs in Altium were regularly solving NP-hard problems, computer science would be a different place. The only goal I see is to meet that bar without hiring layout techs.


>Traces being, say 20% longer than they need to be doesn't really change the performance of the board

Impedance matching requires you know what you are doing, and Altium will only take you so far. For the price, Altium/Protel has always been a poor deal for what you get software wise.

Even free Qspice (a more modern Analog Devices LTSpice) is a far better option:

https://www.qorvo.com/design-hub/calculators-simulation/qspi...

>the industry doesn't look for optimal, only good enough

Indeed, that is why your iPhone still weighs 3.7 kg. =3


> Impedance matching requires you know what you are doing, and Altium will only take you so far. For the price, Altium/Protel has always been a poor deal for what you get software wise.

Pretty much every high end board is designed under Altium. QSpice is a different tool for a different purpose.

And like I said, those few cases were the length actually matters tend to be heavily assisted with automation as it is.

> Indeed, that is why your iPhone still weighs 3.7 kg. =3

I guarantee you that an iPhone board isn't an optimal solution, simply good enough.


>I guarantee you that an iPhone board isn't an optimal solution,

DFM almost certainly guarantees an iPhone is both form, and cost optimized. Jobs was very clear early on in the product development that the EE work had to be miniaturized due to the energy density possible setting volumetric minimums on the battery technology at that time.

> simply good enough.

A common philosophy for people that make cost optimized low-end products. Some people love their 3.7kg phone, as it comes with a stylish shoulder strap. =3


Optimized != optimal

Optimal means there aren't any optimizations possible remaining.

Are you seriously going to suggest that there aren't any potential optimizations left in the iPhone's mainboard?


There are physical limits that are hit with any technology. =3

https://en.wikipedia.org/wiki/Sealioning


Are you sure it wasn't Microsoft doing a rug-pull October 13 on offline Office 2021 pro lifetime license installations for MacOS.

A lot of people are not happy about that stunt. =3


Or people outside the US distrusting US technology? There is a confluence of trends pointing away from US based tech. It will take a crazy amount of work to restore trust.

Sure, in the future everything will have a healthy piss-tint yellow shine. =3

I can imagine the conversation with Studio lawyers about contaminating IP with uncopyrightable generative content liabilities. =3

I'm not an experienced video editor, but I don't think the bulk of video editing involves generating new content on the fly except for narrow cases like adding titles, which is where an AI would easily be able to lean on scripting the existing tooling. There's a ton of grunt work involved in fixing up cuts and figuring out where everything is and so forth.

Resolve has had de-aging filters, and TTS voice synthesis for awhile.

If the chat-bot/spy-ware aren't integrated, than I don't see a problem with LLM plug-in opt-in support. =3


Or any clown with access to Intel Management Engine module signing keys.

If you know, than you know why RISCV was fast-tracked in some places... =3


Unfortunately, "AI" speech to text loggers can run a long time before it ever runs out of memory. Every time you have to update the TV software, a few MiB of your data is sent.

The issue is a lot of the TV traffic is encrypted, and given the other issues reported it likely gets a lot worse for privacy advocates. Air-gaps don't necessarily guarantee anything either. =3


Some C compilers include special interpretations of static and volatile to change how both the code is compiled and linked.

It can be "low-level", but it depends how you define what that means in the age of microcode abstractions.

Most people shouldn't write in C, but that is because their skills belong in Application user space. =3


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: